P138 Prediction model to safely cease anti-TNF therapy in Crohn’s disease: individual patient data meta-analysis (IPD-MA)
Bibliographic record
Abstract
Tools for patient stratification to safely cease anti-TNF therapy in Crohn’s disease (CD) are urgently needed. This IPD-MA aims at development of a predictive diagnostic tool for a personalised approach towards anti-TNF cessation in CD. A systematic literature search was conducted to identify studies investigating the risk of relapse and risk factors in CD patients after anti-TNF therapy cessation by using Medline Ovid, Embase, Cochrane, Web of Science and Google Scholar. Cohort studies with >50 CD patients in remission (clinical or biochemical or endoscopic/radiological) were selected. IPD from the original study databases were used for analysis. Inclusion criteria: luminal CD as indication for anti-TNF therapy, duration of treatment ≥6 months. We associated baseline demographic and clinical data (age, gender, smoking, disease duration, Montreal classification, history of surgical resection, type of anti-TNF medication, concomitant immunosuppressants, corticosteroids prior to cessation and previous anti-TNF therapy) with time to relapse using a Cox model. A prediction model was constructed following the ‘TRIPOD’ statement, with backward selection and p > 0.2 as selection criterion. To investigate the predictive performance internal-external validation was applied. A total of 10 cohort studies were identified, IPD were available from 6 studies. Anti-TNF was withdrawn in 1006 patients, who experienced 474 relapses after a median FU time of 14 months (IQR 8–28). At 1-year relapse rate was 36%, ranging from 24% to 44%. At 2-year relapse rate was 54% (41%-82%). Risk factors for relapse were age (HR 0.98, CI 0.97–0.99), smoking at baseline (HR 1.19 (CI 0.96–1.48), disease duration (HR 1.06, CI 1.03–1.10), disease location (L2) (HR 1.04, CI 0.77–1.41), disease location (L3) (HR 1.25, CI 0.96–1.62), +L4 (HR 1.50, CI 1.00–2.27), type of anti-TNF therapy (adalimumab vs. infliximab) (HR 1.18, CI 0.95–1.48), immunosuppressant use (HR 0.68, CI 0.54–0.85), steroid used 6–12 months prior to cessation (HR 1.24, CI 0.72–2.13), ≥1 anti-TNF therapy in medical history (HR 1.37, CI 1.04–1.80). The prediction model had a discriminative ability with a C-statistic of 0.62 (0.58–0.64). Biochemical parameters of remission (CRP, FC, haemoglobin, leucocytes), anti-TNF trough level and endoscopic data will be added to this preliminary prediction model. The overall risk of relapse in CD patients in remission is 37% within 1 year after anti-TNF cessation. Despite associations between clinical parameters and relapse risk, individualised prediction solely based on clinical parameters remains challenging. Improvement of the discriminative ability of the prediction model may be anticipated after insertion of biochemical and endoscopic data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.024 | 0.040 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.014 | 0.061 |
| Bibliometrics | 0.005 | 0.005 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.004 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".